Which Demographics do LLMs Default to During Annotation?
- URL: http://arxiv.org/abs/2410.08820v2
- Date: Mon, 14 Oct 2024 14:22:40 GMT
- Title: Which Demographics do LLMs Default to During Annotation?
- Authors: Johannes Schäfer, Aidan Combs, Christopher Bagdon, Jiahui Li, Nadine Probol, Lynn Greschner, Sean Papay, Yarik Menchaca Resendiz, Aswathy Velutharambath, Amelie Wührl, Sabine Weber, Roman Klinger,
- Abstract summary: Two research directions developed in the context of using large language models (LLM) for data annotations.
We evaluate which attributes of human annotators LLMs inherently mimic.
We observe notable influences related to gender, race, and age in demographic prompting.
- Score: 9.190535758368567
- License: http://creativecommons.org/licenses/by-sa/4.0/
- Abstract: Demographics and cultural background of annotators influence the labels they assign in text annotation -- for instance, an elderly woman might find it offensive to read a message addressed to a "bro", but a male teenager might find it appropriate. It is therefore important to acknowledge label variations to not under-represent members of a society. Two research directions developed out of this observation in the context of using large language models (LLM) for data annotations, namely (1) studying biases and inherent knowledge of LLMs and (2) injecting diversity in the output by manipulating the prompt with demographic information. We combine these two strands of research and ask the question to which demographics an LLM resorts to when no demographics is given. To answer this question, we evaluate which attributes of human annotators LLMs inherently mimic. Furthermore, we compare non-demographic conditioned prompts and placebo-conditioned prompts (e.g., "you are an annotator who lives in house number 5") to demographics-conditioned prompts ("You are a 45 year old man and an expert on politeness annotation. How do you rate {instance}"). We study these questions for politeness and offensiveness annotations on the POPQUORN data set, a corpus created in a controlled manner to investigate human label variations based on demographics which has not been used for LLM-based analyses so far. We observe notable influences related to gender, race, and age in demographic prompting, which contrasts with previous studies that found no such effects.
Related papers
- Hate Personified: Investigating the role of LLMs in content moderation [64.26243779985393]
For subjective tasks such as hate detection, where people perceive hate differently, the Large Language Model's (LLM) ability to represent diverse groups is unclear.
By including additional context in prompts, we analyze LLM's sensitivity to geographical priming, persona attributes, and numerical information to assess how well the needs of various groups are reflected.
arXiv Detail & Related papers (2024-10-03T16:43:17Z) - Seeds of Stereotypes: A Large-Scale Textual Analysis of Race and Gender Associations with Diseases in Online Sources [1.8259644946867188]
The study analyzed the context in which various diseases are discussed alongside markers of race and gender.
We found that demographic terms are disproportionately associated with specific disease concepts in online texts.
We find widespread disparities in the associations of specific racial and gender terms with the 18 diseases analyzed.
arXiv Detail & Related papers (2024-05-08T13:38:56Z) - White Men Lead, Black Women Help? Benchmarking Language Agency Social Biases in LLMs [58.27353205269664]
Social biases can manifest in language agency.
We introduce the novel Language Agency Bias Evaluation benchmark.
We unveil language agency social biases in 3 recent Large Language Model (LLM)-generated content.
arXiv Detail & Related papers (2024-04-16T12:27:54Z) - What Evidence Do Language Models Find Convincing? [94.90663008214918]
We build a dataset that pairs controversial queries with a series of real-world evidence documents that contain different facts.
We use this dataset to perform sensitivity and counterfactual analyses to explore which text features most affect LLM predictions.
Overall, we find that current models rely heavily on the relevance of a website to the query, while largely ignoring stylistic features that humans find important.
arXiv Detail & Related papers (2024-02-19T02:15:34Z) - Sociodemographic Prompting is Not Yet an Effective Approach for Simulating Subjective Judgments with LLMs [13.744746481528711]
Large Language Models (LLMs) are widely used to simulate human responses across diverse contexts.
We evaluate nine popular LLMs on their ability to understand demographic differences in two subjective judgment tasks: politeness and offensiveness.
We find that in zero-shot settings, most models' predictions for both tasks align more closely with labels from White participants than those from Asian or Black participants.
arXiv Detail & Related papers (2023-11-16T10:02:24Z) - Do LLMs exhibit human-like response biases? A case study in survey
design [66.1850490474361]
We investigate the extent to which large language models (LLMs) reflect human response biases, if at all.
We design a dataset and framework to evaluate whether LLMs exhibit human-like response biases in survey questionnaires.
Our comprehensive evaluation of nine models shows that popular open and commercial LLMs generally fail to reflect human-like behavior.
arXiv Detail & Related papers (2023-11-07T15:40:43Z) - "Kelly is a Warm Person, Joseph is a Role Model": Gender Biases in
LLM-Generated Reference Letters [97.11173801187816]
Large Language Models (LLMs) have recently emerged as an effective tool to assist individuals in writing various types of content.
This paper critically examines gender biases in LLM-generated reference letters.
arXiv Detail & Related papers (2023-10-13T16:12:57Z) - When Do Annotator Demographics Matter? Measuring the Influence of
Annotator Demographics with the POPQUORN Dataset [19.591722115337564]
We show that annotators' background plays a significant role in their judgments.
Backgrounds not previously considered in NLP (e.g., education) are meaningful and should be considered.
Our study suggests that understanding the background of annotators and collecting labels from a demographically balanced pool of crowd workers is important to reduce the bias of datasets.
arXiv Detail & Related papers (2023-06-12T02:26:00Z) - Marked Personas: Using Natural Language Prompts to Measure Stereotypes
in Language Models [33.157279170602784]
We present Marked Personas, a prompt-based method to measure stereotypes in large language models (LLMs)
We find that portrayals generated by GPT-3.5 and GPT-4 contain higher rates of racial stereotypes than human-written portrayals using the same prompts.
An intersectional lens reveals tropes that dominate portrayals of marginalized groups, such as tropicalism and the hypersexualization of minoritized women.
arXiv Detail & Related papers (2023-05-29T16:29:22Z) - Towards Controllable Biases in Language Generation [87.89632038677912]
We develop a method to induce societal biases in generated text when input prompts contain mentions of specific demographic groups.
We analyze two scenarios: 1) inducing negative biases for one demographic and positive biases for another demographic, and 2) equalizing biases between demographics.
arXiv Detail & Related papers (2020-05-01T08:25:11Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.